[Editor's note: this article was written by Charlie Morris in August of 1998. As such, we cannot insure that information enclosed is up to date as of the time you are reading it. Please contact the vendors discussed for updated information on their individual offerings.]

Highly targeted marketing is a lofty goal of Web advertising, and there are several ways in which ad-management packages attempt to serve each ad to the most likely mark. Precise control over targeting is essential for large sites that cover diverse subject areas, and clever targeting can greatly improve click-through rates, or so the story goes.

Page Targeting

You may want to display particular ads only on particular pages, as the ads will be more effective if each is matched to the appropriate content. The better packages all let you select which ads appear on which pages, and also provide some way to group pages of related content.

Day/Time Targeting

The better packages let you specify certain days of the week, as well as specific time periods, that particular ads will run.

IP Targeting

Many packages can target ads based on the various data that can be obtained from the browser, such as domain, operating system, and browser type. By comparing the IP addresses to a database, various types of geographic targeting are possible.

User Behavior Profiling

Some packages can track the path a user takes through a site, and build a "profile" of which pages were visited, which ads were seen, and so forth. This profile is stored as a cookie, so it can be used when a visitor returns to a site. This allows you to ensure that the same visitor doesn't see the same ad twice (or choose to bombard them with the same one over and over!), or serve them certain ads based on which pages they've visited.

Registration-Based Targeting

Sites that have registration forms asking for visitor preferences can use that data for ad targeting. Several packages have a provision for registration-based targeting, and one (AdManager's sister product, Engage) even maintains a database of user information from various sites that use the product.